2010 was a year of recovery, innovation and progress. For publishers, advertisers and agencies alike, the Canadian digital media industry was a rewarding place to be in 2010. With today’s release of the 2010 Canada Digital Year in Review, comScore provides a comprehensive summary of Canadian digital media usage, trends and user demographics across social media, video, search, advertising and mobile platforms over the past year and a look ahead at what to expect in 2011. Below are a few of the highlights from the report:

Social Networking Dominates Display Ad Market
One of the fascinating stories of 2010 was the ever-increasing growth of Facebook across virtually all measures of user behavior (and monetization streams). Using comScore Ad Metrix®, we saw that growth in the Canadian display advertising market was in fact driven by Social Networking sites, which served 37% more display ad impressions in Q4 2010 than Q4 2009. Facebook alone now accounts for 1 in 3 display ad impressions delivered to Canadians. As media dollars continue to move online in 2011, the need to measure the pre and post campaign ad effectiveness of display campaigns is growing, and it is essential to use the right measures for evaluation.

Online Video Engagement Highest Among 18-24 Year Olds
Online video continued to prove its ability to reach all Canadian demographic groups, with persons age 35 and older representing 52% of unique viewers in Q4 2010, according to comScore Video Metrix®. However, 54% of all time spent watching video was by persons under the age of 35. 18-24 year olds were the heaviest online video viewers, watching more than 20 hours of online video per month. As the market continues to mature in 2011, there will likely be an increase in the granularity of publisher’s and advertiser’s public reporting, as agencies demand greater transparency into this fast-growing market.

Mobile Measurement in Canada Coming Soon
As we enter 2011, comScore will expand its measurement of the digital marketing mix in Canada with the introduction of comScore MobiLens™, the first syndicated mobile media measurement service of its kind in Canada. MobiLens will outline the major devices, operators and platforms being used by key demographic groups in Canada, and also rank the top mobile brands being accessed through mobile browsers and applications. MobiLens represents a robust consumer data set for all stakeholders in the Canadian mobile marketplace. Below is an illustration of the smartphone vs. feature phone breakdown in several other countries currently being reported in MobiLens. We’ll soon find out how Canada stacks up against these other markets.

For a full summary of the trends in 2010 and outlook for 2011, please download your copy of the 2010 Canada Digital Year in Review.

comScore Chairman Gian Fulgoni will touch on these findings in his upcoming presentation at the IAB Mixx Canada Spring Conference on Thursday March 10.

We will also host a complimentary webinar on this report in a few weeks. Stay tuned for registration information posted on our webinars page.

This year, the Advertising Research Foundation (ARF) is celebrating its 75 year anniversary. Founded in 1936 by the Association of National Advertisers and the American Association of Advertising Agencies, the ARF leads key industry learning initiatives that increase the contribution of research to better marketing, more effective advertising and profitable organic growth. At the upcoming ARF conference March 21 – 23, I’ll be participating on a panel that will review the past 50 years of advertising and advertising research. The ARF asked me to answer a few relevant questions in advance of the conference (which they have published on the ARF web site) and I thought I’d share those with you here.

What do you see as the most important changes and breakthroughs the advertising industry has undergone in the last 50 years?
Technology has been the catalyst for many of the major breakthroughs we’ve seen in the advertising industry over the past 50 years. The introduction of new technologies has not only dramatically changed the way advertising messages can be communicated but also how marketers can measure the impact of their advertising campaigns.

In the 1980s, for example, the availability of point-of-sale scanner data provided a much-needed solution for Consumer Packaged Goods (CPG) and other industries. For the first time, marketers had the tools needed to quickly and accurately measure the impact of price, promotions and print / TV advertising on brand sales, develop sophisticated market mix models, and link sales lift to various promotional and advertising levers. Prior to scanner data, marketers relied on manual store audits and consumer purchase diaries that were considerably less accurate and timely. During this same period, the People Meter was introduced, providing a much improved solution relative to viewing diaries for measuring people’s TV viewing behavior. All of this undoubtedly had a significant impact on the advertising industry, helping marketers better understand TV audiences as well as the advertising effectiveness of the channel. Ironically, at least in the CPG industry, with the availability of more granular and timely measurement of short term sales response, we also saw the emergence of a far greater reliance on price and promotion spending at the expense of investments in branding advertising.

Finally, and perhaps most importantly, the rise in popularity of the Internet throughout the most recent decade provides yet another example of a technologically-driven change driver in the advertising industry. From the development of sophisticated algorithms that paved the way for the massive search ad market to the ever-evolving display advertising eco-system and powerful targeting capabilities, the Internet has impacted all facets of the advertising industry.

More recently, we’ve seen disruptive technologies pave the way for Internet usage via an array of sophisticated mobile devices, providing yet another compelling new avenue for marketers to communicate with consumers

Which challenges have still not been resolved?
A concern I have for the future of the advertising industry relates to the increasing focus on short term sales via direct response tactics and messaging centered on price and promotion versus longer-term brand building advertising strategies. When it comes to digital advertising, for example, the Internet has certainly proven to be an efficient means of communicating with consumers, but it’s clear that this has come at the expense of strong creative and branding advertising. Many times, display ads will communicate only price-oriented messages, and while this can certainly help to generate immediate sales, one has to wonder at what cost in terms of brand value? Do these short-term tactics simply train consumers to buy on the basis of price alone? This problem is compounded by the continued use of inappropriate metrics such as the click on an ad. While relevant for search advertising, the click on a display ad has been proven to be at best an incomplete, and at worst, a misleading metric. However, the click is fast, easy and inexpensive to compute and, unfortunately, its use persists.

Sadly, the rise of the Internet has also seriously damaged certain media channels -- such as print -- as consumers’ media consumption shifted to online. The future of newspapers and magazines, both of which rely heavily on ad revenue that has seen catastrophic declines, is very uncertain.

What do you see as the most critical topics advertising research will be concerned with in the future?
With the fragmentation of media channels, come increased challenges for measuring multi-media audiences and the effectiveness of advertising across channels. While it is true that the research community has done a good job of figuring out how to measure the separate impact of advertising on TV, on radio, on the Internet and in print, it has yet to fully understand how to measure the combined benefits of advertising across all of these channels. As the industry becomes ever more fragmented, the challenge of doing this accurately and cost-effectively only becomes greater. Undoubtedly of paramount importance is the need to measure TV, Internet and Mobile media consumption on a “single source” basis.

Today, comScore introduced a next generation web analytics product to the market called Digital Analytix. We’re excited about it for a lot of reasons, but mostly because it solves a number of ongoing challenges that we believe have caused pain for the digital community for far too long. Early marketplace feedback confirms our thinking: the platform delivers new value and insight in the increasingly complex digital environment, and in the process, finally addresses the chronic and acute pain points that existing solutions create for web analytics professionals today. Digital Analytix provides the flexible, robust, insightful and convenient solution that analysts have sorely needed.

Highly Flexible and Efficient Reporting Architecture
Many of the complaints voiced by analysts are addressed with Digital Analytix. The most commonly articulated complaint is that existing solutions often lack the flexibility to run reports specific to the needs of an organization. Either the platform architecture relies on pre-aggregated reports that may not deliver the needed insights, or the system provides access to raw data but charges dearly for the privilege of running such reports. Often, a whole different system is required to access the raw data. With the pace of change in digital, it’s impossible to anticipate dimensionality that might be important six months from now. Heck, it might not even be invented yet. And who wants to have to call support every time a new cut of data is required to answer a question?

Powered by the proprietary comScore Atomix technology, Digital Analytix offers access to data in its unaggregated form, enabling the flexibility demanded by web analytics professionals and digital business stakeholders. It enables not only ad hoc analysis and creation of new segments on the fly, but also the ability to create new dimensionality on the fly.

Another important difference in the Atomix approach is the efficiency of the data collection process. Unlike other solutions we efficiently minimize the number of server calls needed, which in turn minimizes the burden on the page performance. For example, instead of collecting each ‘heartbeat’ of a video, Atomix collects all the necessary data with a minimal amount of transmission. The result is a more efficient data collection process without diminishing the quality of the measurement or user experience.

Understand Your Audience
Existing platforms also fail to deliver the mission critical insights into the audience dynamics of a site’s visitor base. The proliferation of mobile devices, the consumption of content and advertising by the same user across disparate media platforms and the simultaneous localization and globalization of content all create a large and growing need to understand and plan against audience segments that are continually becoming more granular and dispersed. Today with Digital Analytix, you can answer questions like: Does conversion decrease with visitor age? Do older people—the ones with the money, in my target—find it more difficult to transact on my site? Did the demographic composition of my site change when I added a new shopping feature? Did it appeal to more women? On a sports site, how does content consumption for registered users compare to non-registered users? Digital Analytix shifts the focus from what ‘cookies’ are doing to the behavior of the people those cookies represent. It’s time to stop thinking just about clicks and start thinking about the person behind those clicks.

There are Demographics… and there are Demographics
Even as other web analytics solutions begin to ostensibly incorporate ‘demographic’ data, a legitimate question arises as to the quality and accuracy of such data. comScore’s demographic reporting is rooted in more than a decade of delivering quality audience-level reporting based on the most advanced and accurate technology in our industry today. If there is one thing we’ve come to know from our years in the internet audience measurement business, it’s this: cookies are not people. Factors such as cookie deletion, usage by the same person from different browsers and devices, and usage of the same machine by multiple people within a household are just some of the complicating factors. As a result, gaining insight into a site’s audience is not as simple as appending cookie-based demographic data to its web analytics data.

Digital Analytix, in contrast, takes a completely different approach to audience-level reporting. The solution builds upon comScore’s Unified Digital Measurement (UDM) methodology, where we’ve taken our 2 million person panel and combined it with billions of server calls from census measurement to produce a ‘best of breed’ audience measurement approach. Since being introduced to the market two years ago, UDM has been adopted by more than 90% of the top 100 U.S. media properties, providing us with coverage on 95% of all machines in the U.S. (international markets are following this trend as we roll out UDM across the world.) The combination of this comprehensive cookie data along with our 2 million person panel has given comScore extensive experience with the complex relationship between cookies and people, and has been the basis for many of the algorithms we now use to make both forms of measurement (audience and web analytics) better. We have found that with the UDM methodology, one form of measurement is essential in informing the other.

Our panel is at the core of our demographic measurement. We go to great lengths to ensure its representativeness, and we weight and project it to accurately represent the online population. We also employ a patented technology that enables us to differentiate between multiple users within a household, ensuring a higher degree of accuracy in assigning the correct audience demographics to each Internet session.

We use cookie-level demographic data to fill in the demographic gaps in Digital Analytix, but first we apply additional methodological rigor. Because our research has shown that cookie level demographic data can be prone to errors, we validate it against multiple sources before applying that information. We have seen, for example, that cookie-based gender data to be incorrect between 20% and 30% of the time, with greater error levels in larger households. Validation against additional sources ensures we’re including only data that has passed the test of convergent validity. In order to account for any bias that results from the limited number of sites from which any one vendor collects data, we also use the panel for calibration and weighting.

While some other web analytics systems are just beginning to tinker in audience data by simply appending data at the cookie or the household level, comScore Digital Analytix has already leapfrogged well beyond these approaches to provide validated, representative, and high-quality demographics.

Integration with Other Platforms
Current solutions also frequently lack the ability to fluidly integrate with other systems. Digital Analytix is an open solution that offers integration with other systems, such as Microsoft Office, so that data can flow elsewhere for additional reporting analysis and functionality. Want to integrate with your chosen email or SEM system? No problem. Want to have data flow through automatically to PowerPoint or Excel so that your CMO can get updates without logging in? Digital Analytix can handle that.

Easy Implementation
But even if an organization feels the chronic, day-to-day pain of their existing platform, the perceived cost of switching to a new platform can seem even higher. After what was undoubtedly a painful implementation that took many weeks or months, the thought of switching web analytics providers can send even the most tech-savvy analyst into a cold sweat. Digital Analytix also alleviates this crucial pain point by allowing publishers to leverage their existing UDM census tags in order to deploy the new system. If you’re tagged for Media Metrix, you’re already tagged for Digital Analytix. Clients can now have one tag that powers both their Audience Measurement and Web Analytics solutions. One tag that is optimized for performance, yet infinitely customizable without ever having to retag or change your implementation. Even without UDM tags, implementation on an average site can be done more quickly and with fewer errors in Digital Analytix due to its efficient tag design and data structure.

Digital Analytix: The Future of Digital Measurement
Are you breathing easier yet? Good. That was our intention. We designed Digital Analytix to be the web analytics solution the industry needs to effectively navigate the digital future.

The rapidly fragmenting digital media environment will only get more complex over time. Organizations need greater simplicity to manage this complexity. The promise of Digital Analytix is that it will enable you to unlock your data, understand your audience, and deliver new insights faster and easier than ever before.

Another innovation, brought to you by comScore.

Yesterday I wrote about comScore’s exciting global launch of Digital Analytix, our next generation web analytics platform that aims to overcome many of the most chronic and acute pain points of web analytics professionals today. Digital Analytix delivers on the promises of greater flexibility, integration, ease of implementation and the groundbreaking ability to understand the demographic profile of your site’s audience.

In addition to all these advantages, Digital Analytix offers perhaps an even more important promise of a better future for digital measurement. This platform provides the answers to one of the most acute pain points our industry has experienced over the years – the debate over which “unique visitor” count is right.

Chronic 'Unique Visitor' Pain Plagues Industry
Historically, web analytics tools have used the terminology of “unique visitors” as a name for a metric that is essentially akin to “unique cookies.” However, people who are not versed in the intricacies of web analytics and the underlying counting rules naturally assume that the metric describes unique people. Digital measurement professionals know the difference between cookies and people and understand the basis for the web analytics unique visitor calculation. However, brand managers who buy advertising, media planners, and even other internal executives are often not aware of the difference.

In web analytics, the metric uses the observance of a new cookie to register a “unique visitor,” while in audience measurement the metric is based on actual people. Over the years, our research has consistently illustrated that cookie-based counting methods often show a “unique visitor” count as high as 2.5x that of measures counting people on the SAME set of machines. This is due to effects such as cookie deletion and the use of multiple browsers to access the same site. The difference is magnified when one takes into account the myriad of access devices used nowadays by the same visitor: multiple PCs at home, a different PC at work, not to mention mobile phones, tablets and other connected devices.

As a result, chronic confusion has prevailed regarding the differences between multiple unique visitor metrics, which are supposed to measure the same site audience. Many in the media industry believe that such confusion undermines advertisers’ confidence in digital measurement and adversely affects their willingness to endorse a medium they perceive as lacking reliable measurement.

It is for this reason that industry organizations such as the IAB and ABCe, both of which have a vital interest in the accuracy, transparency and monetization of digital media, have issued guidelines mandating that metrics based on adjusted unique cookies be called “unique browsers” instead of “unique visitors,” which in their view should be reserved for measurement of unique individuals.

Leading web analytics expert Eric T. Peterson summed it up by saying, “It is about time that we all agreed that ‘unique visitor’ reports coming from census-based [i.e. web site server] technologies frequently have no basis in reality. Further, we should all admit that cookie deletion, cookie blocking, multiple computers, multiple devices, etc. have enough potential to distort the numbers as to render the resulting numbers useless when used to quantify the number of human beings visiting a site or property.”

‘Unique Visitors’ Evolving to ‘Unique Browsers’ in Digital Analytix
We are pleased to announce that Digital Analytix is the first web analytics platform to comply with the aforementioned guidelines in adopting the “unique browser” nomenclature. In addition to supporting industry initiatives, we believe this change will help digital measurement professionals avoid having to explain to bewildered management the difference between wildly diverging metrics both known as the “unique visitor.”

After soliciting input from numerous industry leaders and organizations throughout the web analytics ecosystem, such as ABCi, ABCe, the IAB, and WAA, voices began to coalesce around a single solution to this semantic debate. By and large, the industry recognized that the name needs to change, and the name with the greatest consensus behind it has been “unique browsers.” So that is what we’re going to be calling this metric in Digital Analytix.

Importantly, this metric will remain consistent with the WAA’s definition and interpretation:

The number of inferred individual people (filtered for spiders and robots), within a designated reporting timeframe, with activity consisting of one or more visits to a site. Each individual is counted only once in the unique visitor measure for the reporting period… The deletion of cookies, whether 1st party or 3rd party, will cause unique visitors to be inflated over the actual number of people visiting a site.

Bill Perry, auditing director at ABCi commented “At ABC Interactive, we think it’s important for the online publishing and advertising industry to use common language and definitions to the extent possible. We support the nomenclature change that comScore is implementing to differentiate between unique visitors and browsers. This is in line with the standard used by ABCi in the U.S., ABCe in the UK, the IAB, and many other leading industry organizations.”

Net, support for the idea of a name change is widespread within the industry. Nonetheless, this change was not taken lightly on comScore’s part. It was done with the full understanding that such changes can be uncomfortable, even if done to enable progress and innovation, as is the case here. We recognize that web analytics professionals have become accustomed to their understanding of the “unique visitor” metric and that adopting a new name may encounter some initial resistance. But the industry is calling for this change for good reason, and comScore is proud to lead the way by labeling unique browsers appropriately, while also providing a true person-based audience metric. These two essential metrics, calculated differently and used for different purposes, must work side-by-side to deliver both web analytics and audience measurement insights, rather than in conflict with each other. By finally putting this debate to bed, we hope these changes will help our industry focus on what really matters – using data and analytics to unleash insights and drive value for digital businesses, and providing advertisers with clear, reliable measurement that builds confidence in the digital channel .

In today’s frantic digital media environment (how many times have you heard, “I need it yesterday”?), we often get asked by media planners and sellers, “what’s the comp index?” or “what’s my comp index?” For those who may not be familiar with this term, “comp index” is industry slang for the Composition Index metric.

So what is this thing called Composition Index and why should I give a hoot? (Well, for one, because I’m writing about it, of course!) In its most basic form, Composition Index measures the concentration of a particular target group of consumers on a given website or ad network, compared to the concentration of that target in the total Internet population.

Let’s run through an example together. Oh come on, indulge me… this will be fun, I promise.

Let’s say that twitter.com/jefhack (wink, wink, that’s my Twitter profile) reaches ten people each month and five of these kind folks (or shall I say, “followers”) falls into your target audience. In industry speak, comScore computes a Percentage Composition of 50% (i.e. 5 out of 10) for that target. Now, let’s also say that for the U.S. Internet as a whole, there are 200 million visitors and 50 million of them fall into your target audience. The Total Internet Composition % is 25% (50 million out of 200 million).

To compute the Composition Index for visitors (remember that key piece, “for visitors”, because we’re coming back to that shortly), we divide 25% into 50%, giving us 2, and then multiply that result by 100. So, the Composition Index for the target audience among twitter.com/jefhack followers is 200. Wow, 200! As Steven Tyler would say from the American Idol judging seat, “that’s beautiful, just beautiful”.

But, is it beautiful? Well, yes and no. Oh sure, a 200 Composition Index can be a wonderfully beautiful thing if Steven Tyler says it is, and if it is truly measures what matters for the objective at hand. But what is the objective at hand here? Well typically, media planners and sellers are working together to match up an advertiser’s message with the right audience and the right content with the best combination of target reach and frequency.

So what’s my point? The exercise above assumed that the best way to measure Composition Index in the larger context of matching ads to consumers is to use visitors as the most atomic level of measurement. If this were a conversation about magazine readership or TV viewership, my ever-so-engaging style of writing would quickly come to a stop here. But don’t fret. I think you see where I’m headed… and this job isn’t quite finished.

After all, this is the world of digital media where every event can be measured, reported, and analyzed. Time and time again, we must ask ourselves if there is a better way to be looking at something. Are we analyzing the most meaningful metrics? In this case, there may be something better…

Nestled inside of the comScore Media Metrix planning suite are two reportable measures that go by the names of, % Composition Pages and Composition Index PV. Think of these as analogous to what we just described above, except that their atoms are the pages consumed by the target audience. Why does that matter? Why might this be a better metric in this instance? Because we’re trying to place ads in front of those consumers most likely to purchase a particular product.

Let’s get back to my Twitter profile one last time (I swear that this posting is not solely intended to drive up my number of “followers”). We learned above that five (or 50%) of my profile visitors fell into the target audience. But, what if we could learn that my page was viewed 20 times during that same period, and only five of those page views were seen by the target audience. Hmmm. 5 divided by 20 equals 25% Composition PV. Suddenly, my 50% composition of visitors just dwindled to 25% composition of pages.

Why should you care? Well, we know that ads are placed on pages (or in video streams, but we’ll save that discussion for another day). So, the chances of the target audience seeing the advertiser’s message is directly tied to the number of pages they consume. Therefore, 25% is the more relevant measure in this instance, since we’re interested in understanding exposure to advertising.

So now you ask… that all sounds like it makes sense, but can you prove it? I’m glad you asked, because we sure can.

I took the liberty of tapping into comScore’s Ad Metrix service to answer this final question. Ad Metrix is comScore’s leading competitive intelligence tool for understanding the display advertising market. With Ad Metrix, we can learn just how many ads were actually seen by the target audience and tie that back to both the visitor comps and page comps. Let’s take a look at how this plays out for the audience target of Females Age 25-54 for two sample web properties: Everyday Health and iVillage.

They say that the proof is in the pudding. (I should not be writing this piece before breakfast, clearly).

You’ll notice that for both properties, Females 25-54 make up approximately 40% of the unique audience. And if we just stopped at the % Comp measure for visitors (“UV”), you might assume that there is just a 4 in 10 chance of reaching the specified target with ads. But, if we take it a step further and look at the % Comp Pages measure, the odds improve to about 5 in 10. I’m not a big gambler (realized early on that losing wasn’t fun, especially on a college budget!), but even I know that this is not an insignificant change. Another way to look at it is that you can actually have a 20% greater efficiency (50% divided by 40%) in delivering ads against your desired target simply by looking at a better metric.

Now how can we be sure that this Pages-based measure matters more and is truly the better metric? Take a look at the last column in the table. Because of Ad Metrix, we can quantify exactly how many ads were seen by this target audience on each site (and isn’t that the whole point of this exercise in the end?). I’m also no fancy mathematician, but even a quick glance at the numbers will tell you that the pages comp and ads comp values line up pretty darn close… in fact, strikingly close. The proverbial pudding tells us that the theory of pages consumed indicates likelihood of seeing an ad is quite true.

So my message to media buyers and sellers: Take the extra time to think about whether or not you’re measuring what matters most. And if you’re not, maybe give this PV comp metric a try. It really works!

I hope you enjoyed this ride and will come back again. And oh by the way, my Twitter handle is @jefhack in case you missed that. :-)

In yet another effort to court a younger demographic, Oscar producers picked 29-year-old Anne Hathaway and 33-year-old James Franco to host the famed awards show this year. It was an unusual choice not only because the two are not seasoned comedians—recent hosts have included Ellen DeGeneres, Chris Rock, and Jon Stewart—but also because they are almost 20 years younger than the average age of the telecast’s recent hosts, which has hovered around 50 for the past ten years. If the host selection alone failed to make the effort obvious, the pair’s tongue-in-cheek banter did: “Anne, I must say you look so beautiful and so hip," said Franco in his opening greeting. "Thank you, James," Hathaway quipped, "you look very appealing to a younger demographic as well."

A real demographic concern hides behind this self-effacing humor: the Academy Awards traditionally appeals more to an older film community rather than to younger viewers. But, the goal is now to make the Academy Awards more inviting for younger audiences. The ABC broadcast is the second most-watched TV event after the Super Bowl and drew an impressive 41.2 million viewers last year -- a five-year high. But although the event remains pricey, at $1.7 million a spot, the broadcast's median age in 2009 was 49.5. Last year's ratings among 18- to 34-year-olds actually decreased 3% despite the increase in total viewers.

Some have speculated that the nomination process contributes to the demographics divide. Anecdotally, this seems true. Take 2008, when the wildly successful summer blockbuster ‘The Dark Knight’ was pushed out of the race by films like ‘The Reader’ and ‘Frost/Nixon’. It is easy to imagine indignant twenty-somethings across the United States irked by the fact that Academy voters snubbed what they considered to be a “great” movie simply because it had the misfortune of being a summer release in favor of dramas that neither they nor their friends had ever seen.

In response to this criticism, the Academy decided in 2009 that it would expand the number of Best Picture nominations from five to ten. Academy president Sidney Ganis hinted at the reason for the switch saying, “I would not be telling you the truth if the words ‘Dark Knight’ did not come up.” Summer films have crept into the short list since the change: three in 2009 and four this year.

Will the Academy’s move to make summer blockbusters more competitive result in a younger demographic? comScore data suggests the effort may be having an impact.

comScore AdEffx provides a granular examination of the demographic differences between summer and winter “film intenders,” (defined according to the time of the year when they searched for movie show-times or movies at a local cinema). The data show that summer film intenders skew slightly younger than winter film intenders, with a higher percentage of visitors in the 18-24 year old and 25-34 year old segments:

In light of the age differences, it’s perhaps not surprising that summer film intenders also tend to have lower incomes than their winter peers. While summer intenders are slightly wealthier than the Internet average, winter intenders are even wealthier.

But a silver lining in this income differential actually benefits the silver screen. Despite their lower income skews, summer film intenders devote a larger share of their online transactions (3.8%) to entertainment and movie tickets than winter film intenders (2.5%). As income decreases, share of online transactions allocated to the movie ticket category increases, suggesting that less affluent summer moviegoers are willing to allocate a higher percentage of their discretionary spending to movie entertainment. So despite having lower incomes than their winter counterparts, summer intenders still drive relatively more spending in this particular category. It’s no surprise the movie industry is courting these consumers.

If a younger, summer movie-going demographic is what the Academy wants, then that is what the Academy got—at least online. In order to understand how demographic interest shifted in the past two years, we compared search behavior by age demographic for the month prior to the Oscars in 2011 vs. 2008 using a selection of Oscar-related search terms like “Oscar nominations” and “Best Picture.” The results showed that searchers interested in the 2011 Oscars skewed younger than in 2008.

More specifically, while the younger demographics — particularly 18-24 year olds and 25-34 year olds — showed below average searching interest in the Oscars in 2008, they showed above average interest in 2011. Not only that, but 18–24 year olds exhibited the highest relative interest by far.

So it appears that the Oscar producers were successful in achieving their objective of courting the younger demographics, at least as evidenced by these consumers’ online behavior. With younger audiences accounting for a substantial portion of box office dollars today, it is important to bring these movie-goers into the Oscar fold and support the long-term health of the motion picture industry.

This post was originally published at SearchEngineWatch on March 7, 2011.

A search query, in its essence, is a question requiring an answer. Fortunately the search engines are intelligent enough to find us what we need with most any query, but some inquiries require more specification than just a term. Searchers use modifiers to fish for greater accuracy with their terms all the time, the proverbial long tail, of course. But what happens with the Searchers who are as specific as directly asking a question? With the recent Demand Media (eHow) IPO and the buzz regarding Google’s algorithm changes, my questions suddenly required answers as well.

For the sake of this analysis, let’s define the marketplace as all search terms in the United States that contained at least one of the following words in January 2011 vs. year ago: Who, What, Where, When, Why, and How.

Below is an illustration of the number of searchers and searches on these terms in January:

Each of these question terms tells its own story, so let’s analyze them one by one.

Who

Searchers ask 4 main questions when using the word “who”:

• Who is?
• Who are?
• What was?
• Who were?

As of January 2011, U.S. searchers ran more than 57 million searches containing “who,” an increase of 38% year over year. These searches were run by 21 million people, representing a year over year increase of 18%. Both of these increases outpaced all other question terms on our list…but why?

The simple answer is Kei$ha, Flo Rida, and Akon. Their songs, “We R Who We R” and “Who Dat Girl” both heavily skew the increased results, potentially driving a Q&A publisher down a rabbit hole regarding search analysis. (Content owners should remember that simply analyzing search term sets without detailed review can lead you down the wrong path!)

As for the modifier questions that deliver relevant answers, “Who is” dwarfs the respective competition, accounting for 7.9 million search clicks to websites in January 2011.

Narcissism and/or paranoia also rule the day here, with 4 of the top 20 “who is” search clicks being:

• Who is searching for me?
• Who is looking for me?
• Who is looking at my Facebook profile?
• Who am I? (A bit philosophical if you ask me, but self centered nonetheless)

Otherwise, it’s all about who’s dating who in the celebrity world.

What

“What” is the second largest question modifier used on our list, with 253 million searches by 52 million people in January 2011. Although the total searches are up 35% since last year, the total number of searchers is only up 6%. (As we analyze the rest of our question modifiers, we’ll see that this 6% number is pretty standard across the board - something I will address a bit later).

The main 5 “what” questions are:

• What is?
• What does?
• What are?
• What was?
• What were?

“What is” takes the prize in this category, with 67 million search clicks from their respective searchers. “What does” and “What are” are no slouches though, with 22 and 15 million clicks, respectively.

When using “what” in your search query, the main definition of the term tends to almost always be a noun and directly leads to learning something new.

Top examples include:

• What is refinancing? (Informed home buyer market anyone?)
• What is Taco Bell meat made of? (Topical news stories)
• What is FICA? (January receivers of W2s)
• What is love? (Baby, don’t hurt me)

These questions tend to be the most straightforward as to their meaning and should allow for easier targeting from a search destination perspective than some of the other terms on our list. As soon as you start typing “what”, it’s all about education.

Where

“Where” comes in towards the lower end on volume with 52 million searches and 20 million searchers in January 2011, up 21% and 6% respectively.

“Where” lends itself to contractions a bit more than its related cousins, as the top variations are:

• Where is?
• Where are?
• Wheres? (exactly as typed…texting has destroyed contractions!)
• Where’s?
• Where was?

“Where is” continues our use of “is” as most important phrasing associated with question searching at 8.7 million search clicks.

What’s interesting about “where” searching in January is that it’s top heavy with tax-related items; specifically searchers wondering “where’s my refund.” This may actually represent strong marketing efforts on the part of the IRS this year, since “Where’s My Refund” is their branded area on IRS.gov to allow tax payers a convenient way to receive updates while efficiently addressing the needs of their constituents. The IRS may not be known as the most forward thinking organization, but its commitment to the web, as well as its investment in its campaigns allows for much greater efficiency than we have seen in past years.

Beyond taxes, “where” is an overwhelmingly location-based question (big surprise), with most of the questions asking where they can accomplish a particular task as opposed to finding something (e.g. Where to buy Amazon gift cards.)

When

“When” runs neck-and-neck with “Where” in volume, with 54.8 million searches and 21 million searchers, up 26% and 6% respectively.

The top 4 “when” variants are:

• When is?
• When was?
• When did?
• When does?

“When is” drove 4.4 million clicks in January 2011, with the mass majority of them concerning upcoming holidays and events. Superbowl, Mardi Gras, Easter, and the State of the Union address bubble to the top that month.

With these queries being so time sensitive, this would be a great market for news organizations to review and consider. News publishers of all kinds online can use these queries to better prepare for annual/recurring events with their SEO architecture, since a large volume of searchers will be asking the same questions year over year and these news organizations will inevitably have content related to these items, year after year. Knowing “when” these searches begin to pop up prior to the events can only help with your targeting efforts.

Why

According to most anyone you talk to, “Why” things happen is the most important part. This defines the meaning behind everything that happens in your life…but when it comes to searching it is the smallest area by far, with only 33 million searches and 13 million searchers in January 2011, up 27% and 6% respectively.

“Why” finally breaks our streak of “is” winners with the following Top 5 list:

• Why do?
• Why is?
• Why are?
• Why did?
• Why does?

“Why do” makes up almost 30% of the search clicks for this modifier, with 9.8 million clicks deliver. “Why do” definitely falls into the same category as “what” questions with edification being the main driver, but it is also prominent with searchers that are confused and looking for people with similar problems. The other modifiers on our list are more straightforward with the problem that needs to be solved or the question that needs to be answered. Even “who” related searches tend to be about other people, while “why” questions are centered around and/or include the searcher:

• Why do I hate myself?
• Why am I depressed?
• Why did I get married?
• Why do we dream?

Any publisher out there pushing “why” content should remember that and act accordingly. ”Why” is less about answering a question about the world around you and more about answering questions about you.

How

The big daddy of our question modifiers, “how” related searches making up almost half of the question marketplace with over 431 million searches by 71 million searchers in January 2011, up 27% and 7% respectively.

Without a doubt, our largest question online is “how to” do something:

• How to?
• How do?
• How much?
• How many?
• How does?

“How to” drove over 253 million clicks in January 2011, up 35% since last year and growing every day. While the other modifiers lend themselves to text-based answers quite often, “how to” searching has shown an exceptionally high affinity for video search results. Google, eHow, YouTube, Yahoo! Answers, and others will all deliver you “how to” instructional videos with your results, whether you run this search on a search engine or on a vertical search site.

Some of the most popular searches month in and month out:

• How to tie a tie?
• How to kiss?
• How to get pregnant?
• How to lose weight?
• How to write a resume?

“How to” searching tends to skew towards younger demographics, which makes sense since you have less knowledge about everything the younger your age (how to kiss and other such variants come to mind). That said, “how to” items related to job searching are as collectively large as any “how to” topic at this stage, reflecting the needs of a population in search of work.

Conclusions

Although every search inherently answers a question, those with the above modifiers help define searcher intent in a very particular way. The marketplace is growing and will likely continue to see growth in the coming years. It is important to note, however, the low overall increase of searchers in relation to the large overall increase in searches.

U.S. searchers are averaging over 120 searchers per searcher at this stage, with that number growing double digits every year. Just keep in mind that the searcher population here is not growing so expediently, so the growth in this marketplace is more about searcher engagement than it is to an expanding searcher marketplace. This makes consumer experience on your website, and the ensuing return visitation if satisfied, so important in the Q&A space.

When a 9.0 magnitude earthquake struck Japan off the Tohoku coast on March 11, it triggered tsunami waves of massive proportions and caused devastation to much of the surrounding region, crippling telecommunication infrastructures, immobilizing landlines and Internet access for millions. In the wake of the disaster, mobile phone usage soared as people all over Japan used wireless networks to communicate and seek out information on the safety of others.

Japan has long been one of the world’s most connected populations, with 100.9 million mobile subscribers in December 2010 and more than 75 percent of these subscribers connecting to mobile media. When compared to users in the US and EU5 countries, Japanese users show a higher propensity to access information via apps and mobile browsers. In a society where mobile use has become an integral part of one’s lifestyle, it does not come as a surprise to find mobile use becoming a crucial lifeline in a time of crisis.

To understand the nature of mobile web usage in the wake of the earthquake and tsunami, comScore compared Japan’s fixed Internet and mobile Internet traffic patterns on an hourly basis in the days before and after the quake hit.

In the two days before the quake hit Japan, total Internet traffic in the country followed a predictable weekday pattern, with traffic peaking around noontime and dipping thereafter. On March 11, Friday traffic followed a similar pattern during the first part of the day, but then showed a significant spike in activity following the quake, reflecting a large demand for information following the disaster. In the two subsequent days over the weekend, there is no noontime spike, but Internet traffic is sustained and even surpasses weekday traffic in the evening hours – a pattern that is very atypical of weekend Internet usage.

A look at mobile Internet traffic alone shows even more interesting results, as the increase in weekend traffic is more pronounced coming from mobile phone usage. Here, the effective increase in Internet consumption is more apparent in the days following the quake, with mobile Internet use doubling at peak usage times from the days before the quake. In the days following March 11, mobile Internet use sustains its increased volume throughout the day - not just in the evenings - and reaches its peak late at night on March 13.

PC and Mobile Internet Usage Soars in Comparison to Average
Supporting these findings, the traffic patterns for March 11-13 show an unusual increase when compared to baseline averages for each of those days of the week. To establish a baseline average for Japan Internet traffic on a typical Friday, Saturday, and Sunday, we averaged the amount of traffic observed on each of those days for the four weeks prior to the earthquake. A look at computer-based Internet traffic for those days following the earthquake compared to the baseline shows significant lifts in traffic, peaking with a 48-percent lift vs. average on Saturday, March 12.

Mobile Internet traffic patterns exhibited even more pronounced increases. March 11 showed a 42-percent lift vs. a typical Friday, while March 12 posted a 69-percent lift vs. a typical Saturday. The relatively higher lift in mobile Internet traffic indicates an even greater reliance on mobile devices in the wake of the event, perhaps reflecting the number of people displaced by the crisis or the need to stay abreast of recent events during times in which they would not otherwise be turning to their mobile devices.

Anecdotal information out of Japan support these illustrations of heavier mobile use following the quake, with reports of mobile phone chargers being sold out en masse throughout affected areas on the day of the quake (as stranded commuters had to rely on mobile phones for communication). With landline connectivity unavailable in some areas, there were reports of others relying on VOIP services such as Skype to connect with others to find more information on damage from the earthquake, traffic reports, and other affected areas.

Heat Map Shows Hourly Spikes in Activity in Hours Following Quake
A final illustration shows hourly mobile phone use in Japan in the days before and after the quake. While the pattern of mobile use across the country doesn’t seem to change drastically from a glance, there is a relative increase in the intensity of mobile use in the evening hours in the days following the quake, consistent with what the rest of our data is telling us.

All these data offer a compelling story of how people leverage technology for news, information and communication in the wake of a crisis. But while mobile phones undoubtedly facilitated essential communication and knowledge-sharing during this disaster, it’s important to consider the limitations of this technology, given its dependence on having wireless infrastructure in place.

It will be interesting to see if mobile continues to play a similar role in disaster relief efforts in the future and how the technology could evolve to better serve this need. Similar to how social networks have become an integral part of mobilization in the recent and continuing revolutions in the Middle East, an important question to ponder is whether mobile could eventually become the primary medium for communication in a crisis?

The Journal of Advertising Research (JAR), published by the Advertising Research Foundation (ARF), recently celebrated its 50th anniversary. In that span of time, the JAR has published more than 2,000 articles written by leading academics and practitioners. As part of the 50th anniversary special issue, the JAR asked me to summarize my perspective of Internet-based research over the years with a view towards the future. I thought you might be interested and have included my article below.

The Internet has outpaced every other medium in terms of the speed of consumer adoption. At this writing, more than 1.3 billion people worldwide age 16 or older have online access via fixed ISPs or mobile devices¹. And they use the Internet regularly for communication, entertainment, commerce, and as a timely source of information about a broad variety of issues. In fact, online activities in the U.S. now account for about 25 percent of consumers’ media consumption time².

With such extensive consumer-based online activity, it’s not surprising that marketers are using the medium aggressively as both a sales and marketing channel. This activity also has driven the need for researchers to provide marketers with metrics on how consumers use the Internet. At the same time, the market-research industry has adopted the use of online surveys as a fast, inexpensive way to obtain consumer insights that can be used by marketers in both the online and offline worlds.

The emergence of online research has not been without controversy. Described as “the most measurable medium ever created,” the Internet initially was believed by many to be easily and accurately measured using website servers that recorded every visit to a site or every ad delivered. By placing a cookie (a small piece of computer code) on each visiting computer, it was thought that the number of unique cookies would reveal the number of visitors to any website.

It was only when the market-research industry used using time-tested approaches (continuously tracked behavioral panels, for instance) to show that cookies were being deleted by about 30 percent of all Internet users - and that these so called “deleters” were doing so about four times per month - that it became apparent that counting cookies had led to dramatic overstatements of the actual number of unique visitors to websites. Similarly, it was concluded that using ad servers and placing a cookie every time an ad impression was delivered to a computer in an attempt to measure the reach and frequency of an ad campaign simply led to an overstatement of the reach and an understatement of the frequency actually delivered.

At about the same time, market researchers using behavioral panels were showing that online advertising was not just about direct response and that the click on a display ad (or lack thereof) ignored the latent branding impact of advertising on subsequent brand choice. Put another way, consumers can respond positively to online ads even without clicking on them. This was especially surprising to the many new Internet advertising practitioners who had no previous experience with advertising research. This finding, however, was less surprising to market researchers who had toiled for years in the advertising arena prior to the emergence of the Internet and who understood the many different and complex ways in which advertising can affect consumer behavior.

All of this is not to say that web servers don’t have a place in online market research. Just that one needs to carefully consider what the servers are actually counting. They do indeed capture all of the visits to a website and all of the ad impressions that are delivered. But they also capture “bot traffic” (i.e. computer-driven) site traffic and non-requested pages (sometimes called “push traffic”) which must be filtered out of any visitation metrics, and they are unable to determine which specific individual is using a computer at any point in time. For example, if a single person visits a website from his/her work computer and also from his/her home computer, the website server will count this as two unique visitors, whereas in reality it is one unique person visiting the website twice. Or, if two different people in the same household visit the same website at different times using the same computer, the website server will count this as two visits by one unique visitor whereas in reality it’s two unique visitors. These are obviously important distinctions that any marketer or researcher needs to understand.

Interestingly, leading edge research today is increasingly centered on the use of integrated website server data and continuously measured behavioral panels. By leveraging the strengths of each database and taking into account their respective weaknesses, it’s possible to provide faster, more accurate and more granular insights into online consumer behavior. This is particularly important as the research industry seeks to understand how marketers can best utilize the Internet in a world where social networks are changing the manner in which consumers seek out, receive and use information that affects their brand choice.

On the survey side, substantial research has been conducted by individual survey research companies as well as the ARF to demonstrate that online panels - even if not built according to strict probability sampling rules - can be used to provide fast, accurate and inexpensive insights into consumer attitudes, sentiment and even behavior. Today, research is ongoing to determine the limitations of online survey panels and the situations or applications where probability-based samples are still required.

To a certain degree, what we’ve observed with the emergence of Internet research is the reality that, when attempting to measure a computer-driven medium, one needs to carefully consider what metrics the computer is counting and providing. As Albert Einstein famously said some time ago: “Not everything that can be measured matters.” Ultimately, marketers need to understand the behavior of people and not just computers.

¹ comScore, 2010
² Research conducted by Knowledge Networks for TVB